The epidemiology of personality disorders includes study of their characteristics, prevalence and distribution, and association with psychopathology. It is best conducted with a sample drawn from the general population. A cohort of 900 individuals, originally a probability sample of residents of Eastern Baltimore over 18 years of age in 1981, will be studied. The goals of this research are: (1) to assess the latent structure of personality disorder constructs; (2) to estimate the prevalence of eleven DSM-IV personality disorders and determine the distribution of personality disorder dimensions; (3) to evaluate the relationship between personality disorders and specific Axis I conditions; and (4) to investigate the social costs and consequences of personality pathology. The subjects will be examined by psychiatrists using the SCAN. Personality assessment will be conducted with multiple modalities of measurement including a structured personality disorder interview (PDE), clinical evaluation by a psychiatrist, informant assessment using the PDE, and two self-administered questionnaires which measure different personality dimensions (NEO-PI and TCI). A variety of analytic techniques will be used, including estimation of weights to permit adjustment for attrition by age, sex and race. This current proposal minimizes costs by taking advantages of the tracing and diagnostic evaluations of the ECA follow-up study which is already funded. It provides an unusual opportunity to develop insights into personality disorders; their diagnosis, population rates and distribution, and course and consequences.